Primordial germ cells in an oligochaete annelid are specified according to the birth rank order in the mesodermal teloblast lineage.

The primordial germ cells (PGCs) in the oligochaete annelid Tubifex tubifex are descentants of the mesodermal (M) teloblast and are located in the two midbody segments X and XI in which they serve as germline precursors forming the testicular gonad and the ovarian gonad, respectively. During embryogenesis, vasa-expressing cells (termed presumptive PGCs or pre-PGCs) emerge in a variable set of midbody segments including the genital segments (X and XI); at the end of embryogenesis, pre-PGCs are confined to the genital segments, where they become PGCs in the juvenile. Here, using live imaging of pre-PGCs, we have demonstrated that during Tubifex embryogenesis, pre-PGCs (defined by Vasa expression) stay in segments where they have emerged, suggesting that it is unlikely that pre-PGCs move intersegmentally during embryogenesis. Thus, it is apparent that pre-PGCs derived from the 10th and 11th M teloblast-derived primary m blast cells (designated m10 and m11) that give rise, respectively, to segments X and XI are specified in situ as PGCs and that those born in other segments become undetectable at the end of embryogenesis. To address the mechanisms for this segment-specific development of PGCs, we have performed a set of cell-transplantation experiments as well as cell-ablation experiments. When m10 and m11 that are normally located in the mid region of the embryo were placed in positions near the anterior end of the host embryo, these cells formed two consecutive segments, which exhibited Vasa-positive PGC-like cells at early juvenile stage. This suggests that in terms of PGC generation, the fates of m10 and m11 remain unchanged even if they are placed in ectopic positions along the anteroposterior axis. Nor was the fate of m10 and m11 changed even if mesodermal blast cell chains preceding or succeeding m10 and m11 were absent. In a previous study, it was shown that PGC development in segments X and XI occurs normally in the absence of the overlying ectoderm. All this strongly suggests that irrespective of their surrounding cellular environments, m10 and m11 autonomously generate PGCs. We propose that m10 and m11 are exclusively specified as precursors of PGCs at the time of their birth from the M teloblast and that the M teloblast possesses a developmental program through which the sequence of mesodermal blast cell identities is determined.